Is there an AI bubble? Or are there more than one?
Investing.com -- Deutsche Bank told investors in a recent report that the central market question heading into 2026 is straightforward but difficult: “Is there an AI bubble?”
With valuations surging and the global economy “riding on AI capital expenditure just three years since the launch of ChatGPT,” analyst Adrian Cox said investors, especially those “scarred by the dot-com bubble,” are increasingly uneasy.
The bank argues that even if this is a bubble, “it is still in its early stages.”
Selling now could mean “leaving significant gains on the table,” Cox wrote, noting that the dot-com mania lasted more than three years after Alan Greenspan’s 1996 warning about “irrational exuberance.”
Importantly, Deutsche Bank says history shows there can be multiple bubbles at once.
The dot-com era had two, Cox wrote, with “one with lightly capitalised, never-profitable IPOs and the other with heavily indebted telecoms companies that laid down fibre optic cables that remained dark for years.”
Today’s boom is also said to split. The analyst highlighted the “well-established big tech companies” funding AI data centres from free cash flow, and the “unprofitable companies at the cutting edge,” which remain private and may never fulfil their spending commitments.
Part of the confusion, Cox suggests, is linguistic. He stated that “there is not just one kind of bubble,” and investors should distinguish between valuation, investment and technology bubbles.
On all three, the bank sees green flags, valuations that are “rangebound, earnings-led,” investment that is “cashflow funded,” and technology that is “still scaling, demand-led, [with] falling costs.”
Still, Deutsche Bank warns of risks. Potential red flags include “circular finance valuations,” “debt-fuelled investment,” technological hurdles, political backlash and supply bottlenecks.
The bank concludes: “Reports of a bubble are exaggerated (for now).”
